PSI - Issue 2_A

Akio Uesugi et al. / Procedia Structural Integrity 2 (2016) 1413–1420 Author name / Structural Integrity Procedia 00 (2016) 000 – 000

1415

3

Fig. 3. SEM image of the fabricated specimen. (observing angle: 45°)

among the specimens, so that any effects due to the different orientations would be clarified. The specimens were fabricated on SOI wafers with P-type (boron-doped) device layers whose electrical resistivity was 10 – 30 Ω ·cm. Figure 2 shows schematic of the fabrication process. The fabrication process possessed UV lithography, a deep RIE of the device and handle layers using the Bosch process, and a sacrificial oxide etching. An i-line 5x reduction stepper was used in the UV lithography on the device layer to pattern the specimen with a high resolution and uniformity. Figure 3 shows a scanning electron microscope (SEM) image of the fabricated specimen. Sidewall surface of the specimen exhibited periodic undulations with about 100-nm cycles as scallop damage owing to the Deep RIE and also exhibited a small step of about 50 nm on the middle due to a switch of etching conditions of the Deep RIE. The lower area of the sidewall was etched employing notch-free process to prevent damage on lower edges of the specimens. Figure 4 shows the system for tensile testing at high temperature in a vacuum (Uesugi et al. (2015)). The tensile testing was carried out in a vacuum chamber with a chiller. Tensile force was applied using a piezoelectric stage (SFS-H40X, SIGMAKOKI) and measured with a load cell with a capacity of 500 mN (LTS-50GA, Kyowa Electronic Instruments). The above two instruments were connected to the testing chip and the gripping probe via quartz arms and fixture jigs made of conductive boron nitride whose coefficient of thermal expansion is negligibly small. The testing chips were aligned using accurately milled slots on the fixture jig to control the orientation angles, as shown in Fig. 4b. The displacement of the piezoelectric stage was determined employing a laser displacement sensor (LC-2430, Keyence Corporation) placed adjacent to the chamber, which was not affected by the testing temperature. The testing system possessed an infrared (IR) light heating system IR light from two lamps were concentrated using ellipsoidal mirrors to the center of the chamber where the tensile testing was performed (hereafter, testing area). The concentrated IR light covered the whole testing area to shed temperature non-uniformity in it. The testing area was locally heated in the chamber, because the quartz arms possessed a high IR transmission and a low thermal 2.2. Testing system and conditions

Fig. 4. (a) Schematics of testing machine (Uesugi et al. (2015)); (b) the testing area.